比较SMILES和SELFIES的代币化,用于增强化学语言建模
Miguelangel Leon1, Yuriy Perezhohin1, Fernando Peres1
1NOVA Information Management School (NOVA IMS), Universidade Nova de Lisboa, 1070-312, Lisbon, Portugal.
Scientific reports
|October 24, 2024
概括
用SMILES表示的原子对编码 (APE) 提高了生物物理学和生理学任务中的化学语言模型性能. 这种新的代币化方法可以提高药物发现和材料科学应用的分类准确性.
科学领域:
- 计算化学是一种计算化学.
- 生物信息学是一种生物信息学.
- 在生命科学领域的机器学习.
背景情况:
- 生命科学研究是资源密集型的,通常依赖于试错.
- 机器学习,特别是深度学习,正在加速科学发现.
- 自然语言处理 (NLP) 正在探索化学语言表示.
研究的目的:
- 评估NLP对化学语言表示 (SMILES,SELFIES) 的标记化方法.
- 评估代币化对基于BERT的生物物理和生理学分类模型的影响.
- 将字节对编码 (BPE) 与一种新的原子对编码 (APE) 方法进行比较.
主要方法:
- 使用基于BERT的模型与SMILES和SELFIES表示.
- 应用字节对编码 (BPE) 和原子对编码 (APE) 代码化技术.
- 评估了使用ROC-AUC的HIV,毒理学和血脑屏障透数据集的模型性能.
主要成果:
- 原子对编码 (APE) 显著超过了字节对编码 (BPE).
- 具有SMILES表示的APE在分类任务中表现出卓越的表现.
- 通过APE增强化学元素完整性和上下文关系的保护.
结论:
- 代币化对于有效处理化学语言至关重要.
- 原子对编码 (APE) 为化学NLP任务提供了更强大的方法.
- 精细的代币化技术可以促进药物发现和材料科学.
相关概念视频
Peptide Identification Using Tandem Mass Spectrometry
6.4K
Tandem mass spectrometry, also known as MS/MS or MS2, is an analytical technique that employs two mass analyzers. Essentially it is a series of mass spectrometers that helps isolate a particular biomolecule and then helps study its chemical properties.
This technique helps gather information regarding the protein from which the peptide was obtained and to study the peptides’ amino acid sequence. Identifying peptides from a complex mixture is an important component of the growing field of...
This technique helps gather information regarding the protein from which the peptide was obtained and to study the peptides’ amino acid sequence. Identifying peptides from a complex mixture is an important component of the growing field of...
6.4K
Molecular Models
38.0K
Physical models representing molecular architectures of chemical compounds play essential roles in understanding chemistry. The use of molecular models makes it easier to visualize the structures and shapes of atoms and molecules.
38.0K
Photochemical Electrocyclic Reactions: Stereochemistry
1.8K
The absorption of UV–visible light by conjugated systems causes the promotion of an electron from the ground state to the excited state. Consequently, photochemical electrocyclic reactions proceed via the excited-state HOMO rather than the ground-state HOMO. Since the ground- and excited-state HOMOs have different symmetries, the stereochemical outcome of electrocyclic reactions depends on the mode of activation; i.e., thermal or photochemical.
Selection Rules: Photochemical Activation
Selection Rules: Photochemical Activation
1.8K
SN1 Reaction: Stereochemistry
8.3K
This lesson provides an in-depth discussion of the stereochemical outcomes in an SN1 reaction.
In the first step of an SN1 reaction, the bond between the electrophilic carbon and the leaving group ionizes to generate the carbocation intermediate. The second step of the mechanism is the nucleophilic attack.
In the formed carbocation, the positively charged carbon is sp2 hybridized with a trigonal planar geometry. As all the three substituents lie on the same plane, a plane of symmetry for the...
In the first step of an SN1 reaction, the bond between the electrophilic carbon and the leaving group ionizes to generate the carbocation intermediate. The second step of the mechanism is the nucleophilic attack.
In the formed carbocation, the positively charged carbon is sp2 hybridized with a trigonal planar geometry. As all the three substituents lie on the same plane, a plane of symmetry for the...
8.3K
Chemical Ionization (CI) Mass Spectrometry
705
The molecular ion peak of a molecule in the mass spectrum provides vital information for molecular identification. However, conventional electron impact ionization can lead to the rapid dissociation of some molecular ions before they reach the detector. A milder ionization method is required to increase the lifetime of such ionized analyte molecules. Chemical ionization (CI) is a gas-phase protonation reaction useful for mass-analyzing analyte molecules that are easily protonated to yield the...
705
SN2 Reaction: Stereochemistry
9.3K
In an SN2 reaction, the nucleophilic attack on the substrate and departure of the leaving group occurs simultaneously through a transition state. As the nucleophile approaches the substrate from the back-side, the configuration of the substrate carbon changes from tetrahedral to trigonal bipyramidal and then back to tetrahedral, leading to an inversion in the configuration of the product.
If the substrate is an achiral molecule at the α-carbon, the inversion of configuration is not...
If the substrate is an achiral molecule at the α-carbon, the inversion of configuration is not...
9.3K


